cs.AI（2026-09-18）

📊 共 24 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (15 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (7 🔗1) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (15 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Offline Multimodal Large Language Models for Decision Support in Air Operations	提出离线多模态大语言模型以支持空中作战决策	large language model multimodal
2	MIST: Multimodal Survival Prediction with Genomic-Guided Histology Attention	提出MIST以解决多模态生存预测中的融合挑战	foundation model multimodal	✅
3	PolyBridgeBench: Benchmarking Multimodal LLMs for Physics-Grounded Bridge Design	提出PolyBridgeBench以解决多模态LLMs在桥梁设计中的有效性问题	large language model multimodal
4	LogicTrack: Auditing Reasoning Trajectories of Large Language Models with Formal Logic Solvers	提出LogicTrack以解决大语言模型推理链的逻辑有效性问题	large language model chain-of-thought
5	CESBench: Benchmarking Large Language Models on Cryptographic Engineering Security for IoT Devices	提出CESBench以评估IoT设备的密码工程安全性	large language model
6	Touvigation: Embodied Adaptive Object Acquisition for Blind and Low-Vision Users in Unfamiliar Indoor Environments	提出Touvigation以解决盲人和低视力用户在陌生室内环境中物体获取问题	large language model multimodal
7	LLM-Generated Feature Pools for Time Series Anomaly Detection	提出基于LLM生成特征池的单变量时间序列异常检测方法	foundation model multimodal
8	Verify, Don't Trust: Agentic Model Development for Video Discovery Retrieval at Scale	提出EvoPilot以解决视频检索系统的在线自研究问题	large language model
9	A Lie Detector Test for Language Models: Reading Knowledge a Model Won't Reveal	提出内部识别探测器以解决语言模型知识隐藏问题	large language model
10	Accelerating Dense LLMs via L0-regularized Mixture-of-Experts	提出L0-MoE以加速密集型大语言模型的推理	large language model
11	Samsone: A Family of Open Small Audio Language Models for On-Device Inference	提出Samsone系列小型音频语言模型以实现设备端推理	multimodal
12	One Prompt Does Not Fit All: Self-Meta-Evolve for Personalized Information Extraction	提出Self-Meta-Evolve以解决个性化信息提取问题	large language model
13	Micro-Collaborative Poisoning: A Distributed Attack on RAG Systems	提出微协作中毒攻击以解决RAG系统的安全问题	large language model
14	The Communication Bottleneck: A Round-Trip Study of Tree-Structured Expression Serialization in Language Models	提出回合制协议以解决语言模型中的树结构表达序列化问题	chain-of-thought
15	CogGym: Towards Large-Scale Comparative Evaluation of Human and Machine Cognition	提出CogGym框架以系统比较人类与机器认知能力	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (7 篇)

#	题目	一句话要点	标签	🔗	⭐
16	WS-NeRF: A Mamba-Driven World-State-Aware Adaptive Deblurring Neural Radiance Field	提出WS-NeRF以解决模糊图像对3D重建的影响	Mamba 3D reconstruction NeRF
17	World Modeling in Transformers	提出机制分析以提升变换器的世界建模能力	world model world models affordance
18	Dual-Interest Sequential Product Recommendation With Multi-Granular SSM	提出双兴趣交叉状态空间模型以解决序列推荐中的多义性问题	Mamba SSM state space model
19	GVPO++: Group Variance Policy Optimization for LLM Post-Training and On-Policy Distillation	提出GVPO++以解决LLM后训练和在线蒸馏中的不稳定性问题	distillation large language model
20	GUARD: Natural Forgetting in Large Reasoning Models via Guided Answer-Reasoning Distillation	提出GUARD以解决大型推理模型中的自然遗忘问题	distillation chain-of-thought	✅
21	Calibrating Teacher--Student Discrepancy for On-Policy Distillation	提出Cal-OPD以解决教师与学生模型间的差异问题	distillation privileged information
22	Co-Evolving Zero-Day Jamming: Adaptive Attack Synthesis and Graph Attention-Based Online Detection	提出基于图注意力网络的在线检测框架以解决零日干扰检测问题	reinforcement learning representation learning

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
23	HE-Guardrail: A Homomorphic Guardrail Against Jailbreak Attacks for Encrypted Large Language Model Inference	提出HE-Guardrail以解决HE-LLM推理中的监测漏洞问题	OMOMO large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
24	LEGIT: Credentialing Protocol for Trustworthy AI Agent Marketplaces	提出LEGIT协议以解决AI代理市场的信任问题	manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页